Research On Cross-Layer Design For Wireless Sensor Networks

With the development of hardware and wireless technology, such as embedded-system, sensor technology, Micro-Electro-Mechanical Systems(MEMS), communication, distributed information processing and network technology, which have made the low cost, low power consumption, small size, multi-functional sensor nodes widely developed and applied. Wireless sensor networks(WSNs) are composed of infrastructure-less sensor nodes that have the capabilities of communication, computation and processing information. WSNs are used to collect information from physical world, they sense physical world, process intra-net data, and then transmit the information to the base station by multi-hop. WSNs have significantly impacted on many fields, these applications include military, science, industry, agriculture, medical care and other areas, which redefine the social styles of our life and work.WSNs are large-scale networks deployed hundreds to thousands of nodes powered by batteries. The original network layering structure can’t fulfill the demands of wireless sensor networks, because of their lower reliability and limited resources, the dynamic topology of network and the variable capacity of wireless channel along with the control parameter and environment of the network. Cross-layer design technology that allow interaction and information sharing between different protocol stacks can improve the performance of wireless sensor network system as well as resource utilization, and also can reduce the overhead significantly caused by layered protocol. This dissertation aims to the characteristics of wireless sensor networks, and mainly focus on cross layer design of wireless sensor networks, which includes cross layer design for energy efficient, service of quality and security of networks. The main research work is as follows:1. We analyze the reasons of cross layer design introduced to WSNs, and review the theoretical methods and techniques of cross layer design, and define cross layer design for WSNs. The motivations and advantages of cross layer design for WSNs are analyzed. Cross layer design for WSNs is divided into two categories, namely new abstract design method and compatible cross layer design method, and then evaluate and compare them with performance.2. We proposed a cross-layer QoS routing algorithm which is metric-based cooperative relay initiative forwarding(MCRICQR), in order to meet the QoS of application for WSNs, as well as more resource-efficient for the resource-constrained nodes. We depict a network model, routing model and energy consumption model. An optimal MAC method and MCRICQR algorithm are studied. Timer-based MAC method is used to solve the problem of using the medium and channel based on time-delay sensibility of data that the node sensing. Nodes form a metric according to sensibility, energy, channel, congestion and distance to sink. The node with maximal metric forwards or relays or leapfrogs the data in term of it’s own states, i.e., nodes make a choice for NHF, RFN or LPF to relay packets according to parameters from different layers, such as application layer, network layer, MAC layer and physical layer. They can reduce the transmission delay of packets, improve the reliability and network throughput, achieve load balance, and prolong the lifetime of the network. It has better robustness and adaptability.3. As WSNs have excessive node, widely deployed and dynamic topology, a cross-layer adaptive round distributed cluster-based routing model (CACR) is proposed to achieve more energy efficient. The model consists of two parts:(1) the cross-layer adaptive cluster head election algorithm to elect the cluster head;(2) During data is transmitting to the sink by cluster heads, a cross layer routing of adjacent clusters model algorithm is used. In order to implement energy efficiency for the intra-cluster and inter-clusters communications, the algorithms premeditate the factors, such as residual energy of the nodes, energy consumption factor, RSSI and the distance-energy consumption index. Cluster head work cycle is dynamic adjusted based on residual energy to balance network energy. Transmit power is adjusted to the most suitable level when packet is sending, so as to improve the network energy efficiency, and prolong the lifetime of network.4. A cross layer security framework for WSNs is presented. When a single level of security solution is considered, it exists limitation as the nodes are vulnerable to attack. The security framework is composed of energy-efficient cross-layer security framework and cluster-based secure key management scheme (CSK). And the security framework in energy efficiency, detection and defense against the DoS attack, source authenticity and data validation analysis were carried out. The cross-layer security framework is applied and simulated in WBSNs. In order to implement the security scheme, three algorithms are proposed:(1) Distributed cluster key among cluster heads and their member;(2) Secure algorithms transmitting data to sink node and (3) the security algorithm on sink-side. We evaluated the algorithms. Simulation results show that the secure framework and key management scheme is energy efficient, and they can prevent against stacks of malicious nodes.